Standardization with jupyter/docker-stacks

[sharpTrick]

jupyter/docker-stacks has done an excellent job of creating base jupyter docker images.

They have an clear documentation on Community Stacks which I believe this repo (or its replacement) would be an perfect candidate for.

I am actually working my way backwards from there. I attempted to use the jupyterhub in tandem with its related dockerspawner with docker images this repo alludes to but was unsuccessful.

After stumbling accross this stackoverflow I set out to make my own dockerfile. After achieving initially elusive success I followed the Community Stacks guide to create:

https://github.com/sharpTrick/sage-notebook

My sage installation/implementation is a little different than the strategy used in here. I instead chose to fully rely on the conda implementation. I did this simply because that's how jupyter/docker-stacks base images were implemented. This may or may not be optimal, but it works.

Implementation aside, I believe that building sage's base image (https://github.com/sagemath/docker-images/commit/9cac8785c09c85880f51241638de2ec6c145800e) off of jupyter/minimal-notebook and getting listed as a Community Stack will go a long ways towards reducing fragmentation and achieving some of the goals stated in issue #4.

I realize my implementation may not achieve the goals of this repo, but if by chance they do I will humbly transfer and redirect all my placeholders to their rightful stewards:

• github
• dockerhub
• stackoverflow plug
• jupyter community stack link <- may still be waiting on pull request

Either way the stated goal of this issue is to standardize and defragment via submission to the accepted standards set by https://github.com/jupyter/docker-stacks.